Next Article in Journal
Rectifier Fault Diagnosis Using LTSA Optimization High-Dimensional Energy Entropy Feature
Previous Article in Journal
Three-Channel Fully Integrated Galvanic Isolation Interface in GaN Technology
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

RFID Antenna Deployment Model and Anti-Collision Algorithms to Enhance Vehicle Toll System Efficiency

1
Department of Electronics and Communication Engineering, National Institute of Technology, Aizwal 796012, Mizoram, India
2
School of Computer Science Engineering, VIT-AP University, Amaravati 522237, Andhra Pradesh, India
3
School of Electronics Engineering, VIT-AP University, Amaravati 522237, Andhra Pradesh, India
*
Authors to whom correspondence should be addressed.
Electronics 2025, 14(7), 1404; https://doi.org/10.3390/electronics14071404
Submission received: 27 February 2025 / Revised: 25 March 2025 / Accepted: 27 March 2025 / Published: 31 March 2025

Abstract

:
This paper proposes a standard applications of Radio Frequency Identification (RFID) system to manage vehicles at electronic toll gate system without compromising the vehicle speed. The existing electronic toll gate system has low latency as each car needs to wait in a queue for passing the gate. Installation of reader antenna at each lane makes the system expensive. We propose a two-ray ground model for using minimal reader antenna to define required antenna height and downtilt angle to cover the total gate area. We estimate that reader antenna at a height of 15 m and 45.4° downtilit angle are required for optimal signal reception from the car at a distance of 35–38 m. Modified Pulse Protocol algorithm is suggested for mitigating reader collision and increase throughput and utilization of the system by 14.71% and 15.2% respectively. The probability distribution function of the contention window is modified to triangular function. The M-ary Trimming Q-tree protocol is suggested to resolve the tag collision by trimming the idle nodes from the query tree, which reduces identification time. The total time slot decreases with higher elements, and the total throughput also increases to 98%. The “no-lane” strategy established a fast, efficient and cost effective vehicle toll collection system.

1. Introduction

Radio Frequency Identification (RFID) is one of the most preferred auto-ID technologies among bar code, QR code etc. The word RFID illustrates that object is to be identified by transferring electromagnetic waves. Identification, monitoring and authentication are the significant features of the RFID. A variety of industries have applied RFID technology, from health department to inventory management to industry [1]. This phenomenon is also rapidly implementing to vehicle management over the decade. Parking lot, speed monitoring to electronic toll gate RFID is efficiently employed. Electronic toll collection (ETC) is one of the busiest and most economically beneficial sectors of vehicle transportation. Around the world, RFID is used in almost all ETC. Currently, there is a lane system where each lane has its own reader antenna and tolling booth. The system is therefore expensive to install since each reader has to be paired with a tolling booth. Each lane permits one car to pass the gate at a time hence there is delay to proceed the whole operation.
Therefore, there must have a suitable model of ETC that operates fast and effortlessly at low cost. Figure 1 depicted the outline of the suggested model considering required parameters and corresponding possible issues.
The proposed scenario for ETC is designed in such a way that two reader antennas (no. of antenna can be varied according to the toll gate area) installed at the top of the toll gate. The first and foremost concern is that total number of reader antennas can cover the total gate area. A variety of factors influence the installation mechanism, such as the height of the antenna position, the inclination angle, and the beamwidth. The red colored triangle and circle shape represent the possible issues. The tringle one depicts the reader collision, while two reader antennas are try to read the same tag attached on vehicle. The circle one represents the tag collision, when multiple vehicle respond to the same reader. Corresponding solutions are suggested to mitigate the problems. Modified pulse protocol is proposed for reader collision and M-ry tree algorithm for tag collision.
The position of the reader antenna needs to account the variables such height, downtilt angle, and beamwidth to reduce the path loss and increase the signal strength. Two ray model is the best way to determine the signal strength [2,3,4]. Downtilt angle and beamwidth are required for maximize the coverage area [5,6]. More than one reader antenna within operating region leads to reader collision. RFID functioning is affected in terms of inaccurate and slow readings due to this collisions. To prevent collisions, two mechanisms have been investigated: centralized and distributed. In the centralized method, a central server is responsible for synchronizing all readers and sharing resources. By using the distributed method, readers trade the required information directly to each other without the intervention of a central server [7]. The two methods are based on different protocols, such as the Class 1 Generation 2 UHF standard, Colorwave, and ETSI EN 302 208 [8,9].
On the other side, responding multiple tags at the same time creates tag collision leads to inaccurate vehicle identification. ALOHA-based and tree-based algorithms are used to prevent tag collisions [10].

2. Related Work

There is no research done yet to propose a complete model of RFID implementation in vehicle management at toll gate with no-lane concept. The proposed model has three parts: antenna installation, algorithms to avoid reader collision as well as tag collision. Contemporary research is considered and discussed here that can highlight the work done on each part distinctively.
In [2], two ray model is used to investigate path loss effect of RFID system at different heights. Multi ray model is designed for line of sight communication (LoS) communication at mm wave frequency [4]. Frequency diversity is introduced in two ray propagation model for indoor and outdoor environment [3]. In [11], research revealed that when down tilting mechanism is used, a vertical beam focuses on the vehicle on the road and reduces interference outside of the beam. It is necessary to determine the downtilt angle of the antenna element in order to maximize the coverage range of the reader antenna. Mechanical downtilt angle (MDT) and electrical downtilt angle (EDT) are two common approaches. For mechanical methods, a site visit is necessary to calibrate the tilt angle. The electrical method, on the other hand, does not require site visit, but requires a control system and advanced technology to adjust the tilt angle when necessary [12]. As a result, MDT is the preferred method for simplicity and low cost.
Thereafter, to mitigate reader collision multi-access methods are used which include frequency division multiple access (FDMA), time division multiple access (TDMA), code division multiple access (CDMA), and carrier sense multiple access (CSMA) [13,14,15]. In Class 1 Generation 2 UHF standard, FDMA uses two separate frequency channels to establish communication with tags [16]. Although this method reduces interference between readers, reader-to-tag collisions remain a problem. Furthermore, conventional passive tags do not have any on-board frequency tunning circuit, so they cannot identify specific readers of corresponding frequency channels. TDMA based Colorwave provides different time slots to interfering readers to avoid simultaneous transmission [17]. Each reader selects a different time slot or color to transmit. In the event that any adjacent reader chooses the same time slot, it reselects a new time slot. It is crucial for this method to ensure accurate time synchronization between participants. In CDMA, all tags must be code-designated and need extra on-board circuit, which makes the system complicated and expensive [16,17]. ETSI EN 302 208 is one of the most advanced protocols for RFID applications. CSMA mechanism called ‘Listen before Talk’ is used in the above protocol. If the data channel is idle, the reader starts reading if it is not, it waits for random backoff time [13,18].
Another distributed method, pulse protocol used for passive tags and the method can be made cost-effective. Pulse protocol uses two independent channels. The first is a control channel, and the second is a data channel. For reader-tag communication, a data channel is used. An investigation of the reader’s action is conducted through the control channel [19,20]. As the number of readers increases, collisions persist.
Next, tag collision is reduced by probabilistic or deterministic approach. ALOHA based algorithms could reduce tag collisions but may take a long time to identify the tag, resulting in tag deprivation [10,21]. The tree algorithm is a deterministic approach to receiving tag responses. By enquiring with the identifying number in the read cycle, the reader validates tags in the interrogation zone. There are three types of tree algorithms including collision tree, query tree, and binary tree [10].
A query tree protocol enquires with a series of prefixes. Tag compares the prefix with the tag ID and if the comparison is true, tag responds with the tag ID. A collision mode reader sends a 1-bit prefix query and splits the group into two subgroups. As a result, one subgroup sends the ID to the reader while another subgroup waits for the next query. The reader keeps splitting the group until all tags have been identified [22].
The bit tracking algorithm is a binary procedure. Manchester code is used to detect collided bits. A Manchester code expresses bit utility by following the change in level at transitions [23,24].
A collision tree combines binary trees and query trees by splitting tags into groups and generating a new prefix for the collided tags [25]. Identifying trees and resolving idle cycles are the main challenges of these protocols.

2.1. Comparison with Related Work

In literature, most research has been focused on individual RFID system challenges such as antenna deployment, reader collisions, or tag collisions. All these challenges have been addressed and resolved in our research work. The proposed two-ray solution provides the optimization of antenna height and signal strength accounting the reflection coefficient of the ground path. In outdoor environment, concrete road is considered to determine the received power. Distributed algorithm, Pulse protocol is modified further to reduce reader collision probability and provide equal opportunity to all readers. Query tree is suggested with M-elements and trimming method is also applied to omit the idle nodes for enhancing the identification time of tag.
Antenna propagation model: Comparing the received signal strength, refs. [2,3,4] investigated on general wireless communication model, where, our proposal is for real application of RFID toll system. Instead of approaching multipath propagation like [4] our approach is simple ground reflection phenomenon considering real concrete road having fundamental computational process. Our approach is cost effective on a larger scale as it has not included frequency diversity like [3]. Furthermore, we have estimated the antenna installation specification such as height and downtilt angle.
Reader Collision: Unlike [13], our approach does not need a time slot allocation and can avoid delay in identification speed. There are no messaging scheme between readers in our protocol like [15] which leads to add extra communication system. Sharing information of neighboring reader’s signal strength may cause congestion in [16]. Waiting time interval and probability distribution function of contention window are improved than [20] to reduce collision and fair chance to each reader to communicate with tag.
Tag collision: The prime challenge of [21,22] is to handle large number of tags as the identification method is slot based. In [21], tags are pre-defined correlated ID’s therefore, if tag ID’s are assigned randomly such as vehicle management deployment, efficiency of the algorithm is reduced. In the same manner if ID distribution is not uniform collision may still occur in [10]. Our algorithm is scalable for large number of tags. In [23], multiple tag respond in same slot, increases complexity. In our approach each tag respond when query occurs, reducing interferences. In a real application such as ETC, where large number of random vehicle needed to manage, our approach is more reliable and efficient.

2.2. Motivation of the Proposed Work

The existing ETC model is lane based. Single reader antenna is installed in each lane along with toll booth and system. Vehicles need to stand in queue to pass the toll gate. The following limitations are remain there:
  • Installation and maintenance cost: Installing each reader antenna and maintaining them with system software made the system expensive. Further, it required man power for each booth to operate the system.
  • Throughput and delay: One vehicle at a time creates traffic, cause delay and reduction of the system throughput.
  • Anti-collision algorithm: Suitable anti-collision algorithm is required to identify all the vehicles within less time and without any error.

2.3. Key Contributions

This paper introduces a complete model to reshape the existing ETC structure. It suggested that placement of two reader antenna at a particular height and proper angle is adequate to cover the toll gate area. This way it offers a cost effective, low latency and higher throughput system. It also identified the challenges such as possible reader collision, tag collision and provided required solution. A three-fold solution is proposed based on these objectives:
  • Two-ray model: The installation of the reader antenna makes sure the maximum signal strength to detect vehicle. The concrete road is accounted for the ground’s reflection coefficient when determining the reader antenna’s position. This method helps determine the optimal antenna height and downtilt angle for optimizing signal strength.
  • Modified pulse protocol: A triangular distribution function for the contention window is implemented in the Modified Pulse Protocol algorithm to identify the reader collision challenges and enhance the efficiency of the conventional pulse protocol.
  • M-ary Trimming Query tree: To resolve tag collisions for managing the framework that involved multiple tagged vehicles, the M-ary Trimming Query tree algorithm is introduced. An improved tree-splitting method is improvised with a clipping principle to reduce idle node times, emerge low tag identification times, and increase system efficiency.
Using modified algorithms, we have developed a intrgated model for RFID antenna deployment that includes solutions for reader-tag collisions. In order to verify the analysis and confirm the model’s efficacy, MATLAB version 20.0 is used.
The rest of this paper is organized as follows. In Section 3, the optimal location of the reader antenna in the Two-ray model is comprehensively discussed. As discussed in Section 4, the Modified Pulse Protocol algorithm improves the distribution of contention window to mitigate reader collision challenges. By eliminating idle cycles in Section 5, we are able to construct an M-ary Query tree algorithm that reduces tag collisions and enhances system performance. Section 6 summarizes all executed results. Practical applicability of the proposed research is discuss in Section 7. Finally, Section 8 draws the conclusions.

3. Geometrical Model of Reader Antenna Deployment

According to our proposed model there two or more reader antennas are installed at the toll gate, for simplification we optimized the signal strength of single reader antenna in this section. This is an open unbounded environment with very less probability of having multipath reflection from any scattering object. Therefore, line of sight (LoS) and ground reflection paths are the significant parameters to characterize the propagation model in the reader-tag channel. Hence, two-ray model has been implemented without any loss of generality. Further, both the reader and tag antennas are considered to be polarization matched for the sake of simplicity. Figure 2 illustrates a geometrical model of reader antenna placement in relation to road based on the two-ray ground reflection phenomenon. An evaluation of the effective signal received by the tag on the car is based on the line-of-sight signal (direct wave) from the reader antenna to the car and the reflected wave from the ground. Using Friis’ transmission equation, the power transmitted from the reader to the tag antenna is given as follows [26].
P r T = P t R G t R G r T λ 4 π d 2
where, P t R , G t R are the transmitted power and gain of the reader antenna, respectively. P r T is the received power by the tag antenna and G r T is the gain of the tag antenna, respectively. The distance between reader and tag is d and wavelength of the signal is λ . From (1) it is noted that received power is varies with 1 d 2 .
Referring to Figure 1, the distances of the path travelled by the direct and reflected waves are d1 and d2, respectively. These distances are obtained as follow [3].
d 1 = d 2 + ( H t H r ) 2
d 2 = d 2 + ( H t + H r ) 2
where H t is the height of the reader antenna from the ground, and H r is highest vehicle height.
Considering both the direct and reflected waves, the total received power P T o t a l is given as follows [26].
P T o t a l = P t R G t R G r T λ 4 π 2 e j 2 π λ d 1 d 1 + Γ e j 2 π λ d 2 d 2 2
where Γ is the reflection coefficient of the road which is assumed to be made of asphalt concrete. The value of Γ can be computed from the following equations [27].
Γ = η 2 η 1 η 2 + η 1
Γ = 1 ε 2 1 + ε 2
where η 2 is the wave impedance of the ground and η i is the wave impedance of the air. η i = 377 μ i ε i , where μ i and ε i are the relative permeability and relative permittivity of the ith medium and 377 Ω is the impedance of free space. In this scenario, μ i is taken as unity. The relative permittivity of the Asphaltic concrete material is 3.8 [28]. Therefore, Equation (4) can be written as follows.
e j 2 π λ d 1 d 1 + Γ e j 2 π λ d 2 d 2 2 = C o s 2 π λ d 1 j S i n 2 π λ d 2 d 1 + Γ   C o s 2 π λ d 1 j S i n 2 π λ d 2 d 2
e j 2 π λ d 1 d 1 + Γ e j 2 π λ d 2 d 2 2 = 1 d 1 2 + Γ 2 d 2 2 + 2 Γ d 1 d 2   C o s ( θ 1 θ 2 ) .
where θ 1 = 2 π λ d 1   a n d   θ 2 = 2 π λ d 2 .
The path difference between the direct and reflected waves is denoted as d , which is given as follows [3,4].
d = d 2 d 1
d = d 2 + ( H t + H r ) 2 d 2 + ( H t H r ) 2
Using Taylor series expansion, d can be written as follows.
d = 2 ( H t H r ) d
As, d d , the direct wave distance d 1 is nearly equal to total distance ( d d 1 ) , d 2 = d + d 1 . Therefore, Equation (7) can be rewritten as follows.
e j 2 π λ d 1 d 1 + Γ e j 2 π λ d 2 d 2 2 = 1 d 1 2 + Γ 2 ( d + d 1 ) 2 + 2 Γ d 1 ( d + d 1 )   C o s ( θ 1 θ 2 ) .
e j 2 π λ d 1 d 1 + Γ e j 2 π λ d 2 d 2 2 = 1 d 1 2 [ 1 + Γ 2 1 + d d 1 2 + 2 Γ 1 + d d 1 1 C o s ( θ 1 θ 2 ) ] .
where ( θ 1 θ 2 ) denote the phase difference between the directive wave ( θ 1 ) and the reflected wave ( θ 2 ) . Therefore,
P r T = P t R G t R G r T λ 2 d 2 [ 1 d 1 2 1 + Γ 2 + 2 Γ cos ( θ 1 θ 2 ] .
Conclusion can be drawn from Equation (14) that the power received by the tag antenna is inversely proportional to the square of the ground distance ( d ) and LOS distance ( d 1 ) between the reader antenna and vehicle, when reflected component is present.
The simulation result of received power by the tag at vehicle as obtained from Equation (14) is graphically presented in Figure 3. The simulation is examined with two different antenna heights from the ground ( H t ). It is observed that power response of LOS is monotonically decreased over the distance. In two ray response, the simultaneous occurrence of constructive and destructive interference between the direct ray and reflected ray results in a response with a sequence of maxima and minima. Hence, for a chosen distance ( d ), the preferred height of the reader from the ground is to be determined. For example, if the user wants the tag to be detected at a distance of 35–38 m, then, from Figure 3, it is demonstrated that a height of 15 m is preferred over the other heights, since at that height, maximum constructive interference is taking place at a distance of 35–38 m thereby delivering maximum power with respect to other heights of the reader antenna.
For RFID reader-to-tag communication, the choice of antenna type plays an important role after determining the appropriate height of the reader antenna (15 m).
This proposed system considers the reader antenna as having a directional radiation pattern and the tag antenna as having an omnidirectional radiation pattern. Mechanical downtilt angle is used to estimate coverage area.
The optimum downtilt angle is needed to maximize the area and capacity. It depends on two key features including geometrical factor ( θ g e o ) and antenna vertical beamwidth θ v a r 3 d B . θ g e o is given as follows [29].
θ g e o = tan 1 ( H t H r ) d
Thereafter, the downtilt angle (θ) is determined from,
θ = θ g e o + θ v a r 3 d B 2
Using (15) and (16), the variation of downtilt angle with distance ( d ) is graphically presented in Figure 4. The figure exhibits that the downtilt angle should be decreased with increase in horizontal distance from reader to tag ( d ). Taking the example of Figure 3, if the distance d is desired to be within 35–38 m, the preferred downtilt angle should be 45.4°.
When Reader antennas are tilted downward, the Half-Power beamwidth of the horizontal radiation pattern is reduced in the central direction, but is increased in the sideways direction [5,6]. Figure 5 shows a simulated radiation pattern for a helical reader antenna at various tilt angles. As downtilt angle increases, the gain of the reader antenna decreases and half-power beamwidth (HPBW) increases.
From Figure 5, we can observe that for the example taken, as the downtilt angle is 45.4°, the corresponding HPBW is considerably larger than that for zero downtilt angle. Accordingly, the deployment of adjacent reader antennas after accurately determining the antenna height from ground and downtilt angle will create an interference zone in the reader-tag communication channel as illustrated in Figure 6.

4. Reader Collision

In order to avoid reader collisions effectively, pulse protocol can be applied in a more promising manner. By modifying the waiting state of the reader after the reading operation is completed, the proposed pulse protocol differs from the conventional one. If more than one reader initiates the same random backoff time, the probability of collision increases. As a result, in the proposed method, the waiting time is modified in such a way that the probability of a collision between readers is reduced and the probability of reading the tags increases.
A standard pulse protocol reader transmission is shown in Figure 7a [19], and a modified pulse protocol reader transmission is shown in Figure 7b. Figure 7a illustrates how the reader reads the tag during the time slot as follows: During the period of while communicating with tags, the reader R 1 sends beacon signals to the readers R 2 and R 3 over the control channel. We are unable to read any tags during this time T r e a d . For a period of time T r e a d after transmission ends, all readers enter a waiting state for a period of T m i n . They will resume their backoff time at the end of the time slot.
The backoff time in R 2 is two time slots. A new random backoff time is chosen by R 1 , which is also two time slots. At the end of the backoff time slots of and, both attempt to read the tag simultaneously, resulting in reader collisions. As shown in the figure, within the given time frame, due to uniform waiting time T m i n for all readers irrespective of whether they have communicated with a tag or not and random selection of backoff time, R 2 is not getting a chance to communicate with a tag. Corresponding, algorithm is formulated in terms of modifying waiting time.
As illustrated in Figure 7b, in the proposed algorithm, instead of having uniform waiting time, the waiting time is expanded after the reading slot of each reader. After T r e a d operation of reader R 1 , wait time for R 1 is T m a x instead of T m i n ( T m a x > T m i n ) whereas for the remaining readers R 2 and R 3 , the wait time remains T m i n . Therefore, while R 1 is still in waiting state, R 2 and R 3 start the residual backoff time. As depicted in the figure, the backoff time of R 2 ends prior to R 3 and R 2 starts to communicate with tag. The cycle repeats and R 3 gets the chance to read tag. Hence, this process minimizes the reader.
Collision probability as well as gives all the readers opportunity to communicate with the tags.
The following algorithm describes the graphical representation of the operation. T m a x is the maximum waiting time after successful reading and it is considered as T m a x = 3 T m i n .
The following assumptions are taken to develop the algorithm:
  • Consistently all the readers can establish the communication with tag. But, at a specified time, no more than one reader can communicate.
  • All the readers receive the beacon which implies that there is no hidden reader.
  • One slot size is equivalent to one time unit with the value of beacon interval ( T b i ).
The maximum length of the contention window considered as CW. Reader selects random value from the contention window [0, CW]. Contend backoff time is decreased to zero count and then reader starts communication [20]. In the proposed protocol, the probability distribution of the contention window is considered as monotonically decreasing triangular function. It implies that lesser the backoff time, more is the probability of the reader transmission to tag. The distribution function of the contention window can be expressed mathematically as [30].
p ( x ) = 1 2 C W   C W x ,               0 < x C W
The average of the distribution derives as
C W a v g = 0 C W x   p ( x ) 0 C W x .
Referring to Figure 7, backoff time is decremented after maximum beacon interval, called Adaptive Beacon Interval (ABI). The maximum adaptive beacon interval of the algorithm is T s u c e s s = T r + T m a x + 1. Correspondingly, during the collision period, readers wait for collision slot and next empty slot. Hence, the duration of ABI is T c = c o l l i s i o n   p e r i o d + e m p t y   p e r i o d . Empty period refers to when no collision or successful events take place.
Active ABI is defined when at least one reader communicates. The probability of active ABI is given as [20],
P a c t i v e = 1 1 1 C W a v g N
The probability of successful transmission in ABI duration is given by [20],
P s u c e s s = N C W a v g 1 1 C W a v g N .
Therefore, the probability of collision event in ABI duration is given as [20],
P c o l l i s i o n = 1 1 1 C W a v g N N C W a v g 1 1 C W a v g N .
The total probability of empty slot and collision slot and successful slot is 1. Therefore,
P e m p t y + P s u c e s s +   P c o l l i s i o n = 1
P e m p t y = 1 1 C W a v g N .
Average duration of ABI can be obtained from,
A A B I = P e m p t y T e m p t y + P s u c e s s T s u c e s s + P c o l l i s i o n T c o l l i s i o n .
The probability of collision of R 1 ’s and R 2 ’s transmission is 1 C W a v g . If there are N number of readers, the probability of collision in R 1 ’s transmission is,
P = 1 1 1 C W a v g N 1 .
If collision with probability of P occurs, reader selects another random backoff value from [0, CW] and transmit again at the end the backoff value. Therefore, according to the geometric distribution, the number of transmission trials can be defined with the probability of success ( 1 P ). Thus, until the successful transmission happens the expected number of ABI is defined by [20],
N A B I = 1 P C W 2 + P 1 P 2 C W 2 + P 2 1 P 2 2 C W 2
N A B I = C W 2 ( 1 P ) .
The average duration of reader transmission cycle is given by:
T c y c l e = A A B I × N A B I + T r e a d .
The number of queries is determined by the reader over data channel within T r e a d period. The maximum number of beacon intervals is “a”. Therefore, T r e a d = a × T b i . In each slot of the “a” number of beacon intervals, one beacon is transmitted on the control channel and multiple number of queries on the data channel. T r e a d can be calculated for a beacon interval as [20],
T r e a d = l q + τ q × Q T + a ( l b + τ b ) ,
Q T = T r e a d a ( l b + τ b ) l q + τ q .
Throughput and utilization describe the system’s merit. The number of successful periods by all readers in one cycle is determined by N A B I × P s u c e s s . The number of successful queries sent by reader is Q T × N A B I × P s u c e s s . Therefore, total throughput is defined as [16]
T = Q T × N A B I × P s u c e s s T c y c l e .
Utilization is obtained from,
U = T r e a d × N A B I × P s u c e s s T c y c l e .
The parameters considered in the proposed Algorithm 1 with the corresponding values is tabulated in Table 1.
Algorithm 1: Queue Delay Calibration ()
/* Initialize parameters */
Random   back _ off   ( )   r
Backoff   _ timer   ( )   b t
Waiting _ timer   ( ) w t
Reading _ timer   ( ) r t
Paused   time   ( ) t
if State_value = = 1; /* state is idle*/
w t     T m i n
else if  State_value = = 0 && State = “Contend”
while control channel is busy
b t → ∆t
end while
while control channel is idle
b t     b t + b t
end while
  else if State_value ∉ {0,1} /* state is waiting*/
while   b t > 0
b t     b t b t  
while   b t ≤ 0
b t r
end while
end if
   while control channel is idle
  Beacon is transmitting over control channel
State = “Reading”
end while 
       While   r t ≤ 0
if  State = = “Successful”
       cancel all timer
else if  State = ”Waiting”
     resetting state after reading
w t = T m a x
end while
       end if
     end if
   end if
end if
As can be seen in Figure 8 and Figure 9, the simulation graph depicts the system’s throughput and utilization. Figure 8 and Figure 9 show the responses of the conventional pulse protocol and the modified pulse protocol. When the number of readers is low, the probability of an empty slot is high, and as the number of readers increases, the probability of an empty slot decreases. Therefore, throughput of the system also increases, however large number of readers initiates collision slots. According to the proposed algorithm, the throughput of the system is much higher than conventional one since the average value of contention window and waiting time are modified. With four readers working together in the system, the throughput reached 2165 queries/s for the existing method and 2509 queries/s for the proposed one. As the number of readers increases, communication duration between reader and tag becomes shorter. As a result, the overall utilization of the system reached nearly 92% for the proposed protocol, which is 15.2% higher than the existing protocol.
The execution time analysis is depicted in Figure 10 to define the computational complexity of the proposed Modified Pulse Protocol.
Three main parameters are considered for determining the computational complexity. (i) Contention window as it determines the average back-off time for each reader. (ii) Collision of probability to notify that how many retransmissions are required. (iii) Adaptive Beacon Interval (ABI) as it supervises the duration cycle. Readers are varied from 1 to 20 and execution time is evaluated for each reader.
It is observed from the Figure 10 that execution time is increases in a piece-wise linear manner as number of readers are increased. The maximum execution time is 0.008 s for 20 readers. Therefore, the selection of our contention window as triangle distribution function reduced execution time and established computational efficiency.
Further, memory depends primarily on number of readers and execution time. Therefore, proposed algorithm runs within O(n) complexity. That signify linear memory usage.

5. Tag Collision

By modifying the existing Tree-based algorithm, the proposed communication model resolves the tag collision issue. By reducing idle nodes in the tree, the proposed model increases tag identification time very quickly. A schematic of the query-tree is shown in Figure 11. Based on the tag response, each node in the proposed tree algorithm defines the anti-collision condition. The nodes are categorized into three modes—idle, readable (successful), and collision. These conditions are defined as follows [23,31].
  • Idle: There is no transmission as no tag is present in the query zone.
  • Readable: Successful transmission as single tag responds to the query of reader.
  • Collision: Collision occurs as multiple tags respond to the query of the same reader.
Tag collision can be more efficiently reduced if binary query tree is improved to M-ary query tree. The same number of tags can be identified in M-ary tree algorithm with a less collision cycle however, number of idle cycles increases as compared to binary query tree [22,32]. Hence, to further reduce the idle cycles without compromising the tag collision probability, the M-ary Trimming Query tree is proposed in this work.
As shown in Figure 10, an M-ary tree is divided into M groups of bits, highlighted as b0, b1 up to bm−1. The reader first sends query prefix to tag in communicating zone. The tag responds to the query after comparing with the own ID prefix. Collision bit detection is carried out by Manchester code. If there is collision, reader updates the query prefix and collision node is divided further in the next level into M-groups as illustrated in Figure 10.
The trimming for the M-ary tree method is carried out in three ways. Consider the value of M is 8 and the position of the highest collided bit bm−1 is 0, then all query nodes with a least significant bit (LSB) value 1 are eliminated and rest of nodes are set as new prefix and added to stack. Inversely, if the the highest collided bit bm−1 is 1, then query nodes with LSB value 0 are removed and rest of the nodes are set as new prefix and pushed to stack. If the highest collided bit is unknown then there will be no trimming, it will get forwarded to the subsequent level of the tree.
The key parameter of RFID tag identification is total time slot. In an anticollision algorithm, lesser the total time slot, greater the speed of execution of the algorithm. The analytical calculation is compared between conventional M-ary tree and M-ary trimming tree. Root node, intermediate nodes and leaf nodes are the sections of the M-ary query trimming tree. The level of leaf node is zero and level of intermediate node is equal to level of the tree e.g., M . Therefore, total nodes of the M-ary tree is [33]:
N = n 0 + n M .
where,   n 0 is the number of nodes at level 0 and n M is the number of nodes at level M. The root node is not the child node. Intermediate node has M number of children nodes and leaf nodes don’t have any child node. Thus, total number of nodes of M-ary tree is [33]
N = (   0 × n 0 ) + M × n M + 1 ,
Then,
  n 0 + n M = M n M + 1 .
Hence, number of leaf nodes and intermediate nodes can be determined as,
  n 0 = ( M 1 ) n M + 1 ,
n M =   ( n 0 1 ) ( M 1 ) .
Let, n is the number of tags. The number of leaf nodes is equal to number of tags, which are needed to be identified under the trimming condition as there is no idle slot in this algorithm. Therefore,
  n 0 = n ,
n = ( M 1 ) n M + 1 .
Let, at the trimming condition, the probability of getting least collision at the intermediate node is 1. Therefore, new intermediate node ( n M ) at level M becomes
n M =   ( n 0 1 ) ( M 1 ) × 1 M =   ( n 0 1 ) M ( M 1 ) .
Then time slot ( N 1 ) is,
N 1 = n + n M = n + ( n 1 ) M ( M 1 )
There are possibilities to have M 1 number of idle nodes in a tree and the remaining one is the collision node. Let, n is the readable node and m is the leaf node or idle node. Therefore, time slot ( N 2 ) without trimming condition is also evaluated as,
N 2 = n + m + n M = n + M 1 n M + n M .
N 2 = n + M ( n 1 ) ( M 1 ) .
The M-array trimming tree initiated the total time slot (T) of the algorithm after cutting off the idle node is [33]
T = [ N 1 ,   N 2 ] .
The ratio between the total number of identified tag and required total identification time is defined as throughput of the identification. The efficiency of the algorithm is defined by the throughput ( T h ) [33]. Therefore,
T h = n T .
Figure 12 depicts the comparison between conventional M-ary query tree and proposed M-ary Trimming Query tree by using the values of M = { 2 ,   4 ,   8 } . It is observed from the figure that, for any value of M , the total time slot is reduced in proposed algorithm as it eliminates the idle slots. The figure also shows that as the value of M-arry tree increases the total time slot for tag detection decreases.
Figure 13 depicted the results of system throughput for conventional and proposed algorithm with accounting different M values such as ( M = { 2 ,   4 ,   8 } ). It is clearly seen from the respond that as the element of the tree is increased the identification time is decreased and therefore throughput of the maximum value of ( M = 8 ) tree with trimming capability reached nearly 98% which is 11.8% higher than conventional one with same M value.
To implement the proposed algorithm in real-world application, a middleware or firmware customized RFID reader is needed to communicate with tags. The algorithm can be written within readers software stack through API programming. Tag identification time and responses will register into computer interface. Every time it will record the tag responses and expand the query if tag responses. If there is no response from query branch, it will trim and not query further. This way it will save the query time and enhance the identification time.

Performance Comparison

The performances of the proposed M-ary trimming query tree algorithm (with M = 8) is compared to the binary tree, query tree (QT) and collision tree (CT) protocols. To identify all the tags within fewer time, total time slot defines the effectiveness of the algorithm. Figure 14 displays the total time by varying the number of tags. Binary tree has highest time slot followed by the query tree and collision tree. The proposed tree algorithm has the least time slot compared to other algorithms as it eliminates the idle slot. Correspondingly, throughput of the system is shown in Figure 15 for different algorithms. The figure depicts that the throughput is almost constant over the varying no. of tags for each algorithm. The binary tree attains 15%, whereas QT and CT achieves 50% and 66.6% throughput respectively. The proposed algorithm achieves 98.25% throughput which is significantly higher than the protocols compared. Table 2 highlights the advantage of the proposed algorithm over the existing algorithms in terms of total time slot and throughput.

6. Summary of Results

Our work provides threefold solution for an efficient and cost effective Toll system.
  • Antenna installment: Proposed two-ray model optimized the signal strength considering the concrete ground and provides the solution of antenna installation at the toll gate. A 15 m antenna height and 45.50 tilt angle established the detection efficiency at a distance of (35–38) m.
  • Reader collision: Modified Pulse protocol with modified contention window length ensured fair transmission distribution among the readers and reduced the collision between them. The proposed protocol achieved higher query rates 2509 queries/sec, while minimum four reader are actively working and led to 92% utilization of the system. Further maximum execution time for highest reader 20 is 0.008 s.
  • Tag collision: Proposed M-ary trimming query tree algorithm expanded the queries to M element for reducing identification time and removed no-response node or idle node to enhance the throughput. It achieved 98% throughput, considering M = 8 over the conventional method.
This way it offers a cost effective, low latency and higher throughput ETC model.

7. Practical Implementation

Our proposed work provides a practical solution to enhance the efficiency and scalability of existing ETC system. The key areas of applicability is as follows:
  • Uninterrupted Toll system: This model is applicable for any highway or expressway toll system. The suggestion of eliminating lane system and implementing minimum number of reader antenna for vehicle detection makes the system fast and less congested. No vehicle need to wait in a queue to pass the gate.
  • Multi-reader environment: Proposed Modified Pulse protocol reduce the reader collision while multiple reader cover the toll gate area in no-lane system. Our suggested algorithm enhance the throughput by optimizing reader transmission timing.
  • High dense vehicle management: Number of vehicles need to manage at the highway toll gate causes tag collision. Trimming method of our proposed M-ary trimming query tree algorithm minimize the number of queries and ensure faster tolling system.
  • Cost effective: Our proposed work offers a cost effective toll system. It established a fast, efficient and economical system for managing large amount of vehicles applying minimum arrangements.
  • Future Intelligent Transport system (ITS): The proposed model can integrate to vehicle to vehicle (V2V) or vehicle to Infrastructure (V2I) communication for smart city.

8. Conclusions

The aim of this paper is to propose a systematic approach to reader-tag communication, oblivion of the concept of lanes for fast and efficient toll collection. A road-to-vehicle RFID communication mechanism could be developed to extend the proposed mechanism. There were three parts to the approach: proper positioning of the reader antenna, appropriate algorithms to avoid reader collisions, and tag collisions. The optimum signal strength requires both the height of the antenna and the downtilt angle to be maintained correctly. Cross-over between reader antenna radiation patterns caused the reader collision during this process of signal optimization. Modified Pulse Protocol algorithm avoided this problem by improving throughput and utilization over conventional pulse protocol algorithm. As a result of the proposed algorithm, all readers had an equal opportunity to communicate with the tag and the contention window was distributed triangularly to all readers. Utilizing the M-ary Trimming Query tree algorithm, we were able to resolve the issue of tag collisions by combining the advantages of the M-ary query tree and the trimming method. Trimming reduced idle time slots significantly, which resulted in a significant decrease in the total time required to read a tag as well as increased system performance. In this work, a three-fold mechanism of positioning antenna was proposed as a method for establishing RFID communication between a car tag and a roadside reader. Utilization of appropriate algorithms to eliminate reader and tag collisions enhanced the receiving signal strength, system performance, and utilization, as well as shortened the reading time slot required.
The entire mechanism of no-lane ETC, as presented in the proposed work can be made more efficient in terms of data transmission and security by integrating it with 5G networks. Also, the algorithms can be made AI-driven for optimum management of both tag and reader collisions.

Author Contributions

Methodology, K.M.; Software, K.M. and P.S.; Validation, A.G. and S.P.; Formal analysis, A.G., P.S., S.P. and J.K.; Investigation, K.M.; Resources, A.G. and J.K.; Data curation, A.G. and J.K.; Writing—original draft, K.M.; Writing—review & editing, A.G., P.S., S.P. and J.K.; Visualization, P.S.; Supervision, A.G.; Funding acquisition, S.P. and J.K. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Data Availability Statement

Data are available as needed from the first author.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Mazumder, K.; Ghosh, A.; Bhattacharya, A.; Ahmad, S.; Ghaffar, A.; Hussein, M. Frequency Switchable Global RFID Tag Antennae with Metal Compatibility for Worldwide Vehicle Transportation. Sensors 2023, 23, 3854. [Google Scholar] [CrossRef]
  2. Maeng, S.J.; Kwon, H.; Ozdemir, O.; Güvenç, I. Impact of 3-D Antenna Radiation Pattern in UAV Air-to-Ground Path Loss Modeling and RSRP-Based Localization in Rural Area. IEEE Open J. Antennas Propag. 2023, 4, 1029–1043. [Google Scholar] [CrossRef]
  3. Besser, K.-L.; Jorswieck, E.; Coon, J.P. A Simple Frequency Diversity Scheme for Ultra-Reliable Communications in Ground Reflection Scenarios. arXiv 2022, arXiv:2206.13459. [Google Scholar]
  4. Khawaja, W.; Ozdemir, O.; Erden, F.; Ozturk, E.; Guvenc, I. Multiple ray received power modelling for mmWave indoor and outdoor scenarios. IET Microw. Antennas Propag. 2020, 14, 1825–1836. [Google Scholar] [CrossRef]
  5. Du, L.; Sun, Q.; Cai, C.; Zhang, Y.; Hu, H. Antenna Beamwidths of Above-The-Road Radar for Traffic Speed Enforcement in China. Int. J. Antennas Propag. 2012, 2012, 492326. [Google Scholar] [CrossRef]
  6. Shi, W.; Guo, Y.; Yan, S.; Yu, Y.; Luo, P.; Li, J. Optimizing Directional Reader Antennas Deployment in UHF RFID Localization System by Using a MPCSO Algorithm. IEEE Sens. J. 2018, 18, 5035–5048. [Google Scholar] [CrossRef]
  7. Rezaie, H.; Golsorkhtabaramiri, M.; Navimipour, N.J. A radio frequency identification reader collision avoidance protocol for dense reader environments in the context of Industry 4.0. IET Radar Sonar Navig. 2022, 17, 362–373. [Google Scholar] [CrossRef]
  8. Su, J.; Liu, A.X.; Sheng, Z.; Chen, Y. A Partitioning Approach to RFID Identification. IEEE/ACM Trans. Netw. 2020, 28, 2160–2173. [Google Scholar] [CrossRef]
  9. Golsorkhtabaramiri, M.; Issazadehkojidi, N.; Pouresfehani, N.; Mohammadialamoti, M.; Hosseinzadehsadati, S.M. Comparison of energy consumption for reader anti-collision protocols in dense RFID networks. Wirel. Netw. 2019, 25, 2393–2406. [Google Scholar] [CrossRef]
  10. Su, J.; Sheng, Z.; Huang, C.; Li, G.; Liu, A.X.; Fu, Z. Identifying RFID Tags in Collisions. IEEE/ACM Trans. Netw. 2022, 31, 1507–1520. [Google Scholar] [CrossRef]
  11. Wilson, G. Electrical downtilt through beam-steering versus mechanical downtilt (base station antennas). In Proceedings of the Vehicular Technology Society 42nd VTS Conference–Frontiers of Technology, Denver, CO, USA, 10–13 May 1992. [Google Scholar]
  12. Samal, S.R.; Swain, K.; Bandopadhaya, S.; Dandanov, N.; Poulkov, V.; Routray, S.; Palai, G. Dynamic Coverage Optimization for 5G Ultra-dense Cellular Networks Based on Their User Densities. Wirel. Pers. Commun. 2022, 128, 605–620. [Google Scholar] [CrossRef]
  13. Golsorkhtabaramiri, M.; Issazadehkojidi, N. A Distance Based RFID Reader Collision Avoidance Protocol for Dense Reader Environments. Wirel. Pers. Commun. 2017, 95, 1781–1798. [Google Scholar] [CrossRef]
  14. Cmiljanic, N.; Landaluce, H.; Perallos, A. A Comparison of RFID Anti-Collision Protocols for Tag Identification. Appl. Sci. 2018, 8, 1282. [Google Scholar] [CrossRef]
  15. Mafamane, R.; Mansour, A.A.; Ouadou, M.; Minaoui, K. FTSMAC: A multi-channel hybrid reader collision avoidance protocol for RFID network. J. Sens. Actuator Netw. 2021, 10, 46. [Google Scholar] [CrossRef]
  16. Golsorkhtabaramiri, M.; Tahmasbi, M.; Ansari, S. A Distributed Mobile Reader Collision Avoidance Protocol for Dense RFID Networks. Wirel. Pers. Commun. 2022, 125, 2719–2735. [Google Scholar] [CrossRef]
  17. Loginov, V.; Khorov, E.; Lyakhov, A.; Akyildiz, I.F. CR-LBT: Listen-Before-Talk with Collision Resolution for 5G NR-U Networks. IEEE Trans. Mob. Comput. 2022, 21, 3138–3149. [Google Scholar] [CrossRef]
  18. Pei, X.; Qian, H.; Wang, H.; Kang, K. An Improved Listen-Before-Talk Scheme for Uplink Multiple Access in 5G Unlicensed Band. IEEE Internet Things J. 2022, 9, 19843–19853. [Google Scholar] [CrossRef]
  19. Jiang, Y.; Cheng, W.; He, Y.; Pan, S. A Multi-channel Based Reader Anti-collision Protocol for Dense RFID Environments. In IoT as a Service. IoTaaS 2018; Lecture Notes of the Institute for Computer Sciences, Social In-formatics and Telecommunications Engineering; Li, B., Yang, M., Yuan, H., Yan, Z., Eds.; Springer: Cham, Switzerland, 2019; Volume 271, pp. 384–392. [Google Scholar]
  20. Birari, S.M. Mitigating the Reader Collision Problem in RFID Networks with Mobile Readers. Master’s Thesis, Computer Science & Technology, Indian Institute of Technology Bombay, Mumbai, India, 2005. [Google Scholar]
  21. Xie, X.; Liu, X.; Qi, H.; Guo, S.; Li, K. A Tag-Correlation-Based Approach to Fast Identification of Group Tags. IEEE Trans. Mob. Comput. 2021, 21, 3096–3109. [Google Scholar] [CrossRef]
  22. Yang, X.; Liu, X.; Wu, B. Smart Batch-Processing Protocol for RFID Multiple Tag Identification. IEEE Access 2020, 8, 159132–159142. [Google Scholar] [CrossRef]
  23. Su, J.; Chen, Y.; Sheng, Z.; Huang, Z.; Liu, A.X. From M-Ary Query to Bit Query: A New Strategy for Efficient Large-Scale RFID Identification. IEEE Trans. Commun. 2020, 68, 2381–2393. [Google Scholar] [CrossRef]
  24. Guo, K.; Xie, X.; Chen, S.; Qi, H.; Li, K. Efficient collision-slot utilization for missing tags identification in RFID system. Comput. Commun. 2022, 195, 61–72. [Google Scholar] [CrossRef]
  25. Song, J.; He, S.; Yao, H. TMIA: A Tree-Based Multi-Reader Interactive Anti-Collision Algorithm for RFID Tag Identification. IEEE Access 2020, 8, 81594–81605. [Google Scholar] [CrossRef]
  26. Mudhigollam, U.K.; Mandava, M.R. An analytical study of wireless power transmission system with metamaterials. Energy Harvest. Syst. 2023, 11, 20220135. [Google Scholar] [CrossRef]
  27. Pozar, D.M. Microwave Engineering, 4th ed.; Wiley: New York, NY, USA, 1998. [Google Scholar]
  28. Zhu, T.; Carcione, J.M.; Botelho, M.A. Reverse time imaging of ground-penetrating radar and SH-seismic data including the effects of wave loss. Geophysics 2016, 81, H21–H32. [Google Scholar] [CrossRef]
  29. Blitti, K.; Dotche, K.A.; Diawuo, K. Macro-cell antenna down-tilt to field data study in Ghana. In Proceedings of the 2012 IEEE 4th International Con-ference on Adaptive Science & Technology (ICAST), Kumasi, Ghana, 25–27 October 2012. [Google Scholar]
  30. Taub, H.; Schilling, D.L.; Saha, G. Principles of Communication Systems, 4th ed.; McGraw-Hill: New York, NY, USA, 2017. [Google Scholar]
  31. Lin, C.-C.; Lai, Y.-C.; Hailemariam, Z.L. A distinguished-bit tracking knowledge-based query tree for RFID tag identification. Comput. Commun. 2024, 218, 166–175. [Google Scholar] [CrossRef]
  32. Nguyen, C.T.; Hoang, L.T.; Dang, K.D.; Pham, A.T. Enhanced M-Ary Detecting Tree: A Novel Time-Efficient Protocol for Tag Identification in Active RFID Systems. IEEE Access 2024, 12, 39517–39532. [Google Scholar] [CrossRef]
  33. Zhang, W.; Guo, Y.; Tang, X.; Cui, G.; Wu, L.; Mei, Y. An Efficient Adaptive Anticollision Algorithm Based on 4-Ary Pruning Query Tree. Int. J. Distrib. Sens. Netw. 2013, 9, 848746. [Google Scholar] [CrossRef]
Figure 1. The outline of the complete model of the suggested method.
Figure 1. The outline of the complete model of the suggested method.
Electronics 14 01404 g001
Figure 2. Layout of the proposed two-ray ground model.
Figure 2. Layout of the proposed two-ray ground model.
Electronics 14 01404 g002
Figure 3. Maximum strength of two ray transmission at two different heights.
Figure 3. Maximum strength of two ray transmission at two different heights.
Electronics 14 01404 g003
Figure 4. Downtilt angle determination with preferred height of reader antenna.
Figure 4. Downtilt angle determination with preferred height of reader antenna.
Electronics 14 01404 g004
Figure 5. Radiation pattern for three different tilt angles (0°, 45.4°, 90°).
Figure 5. Radiation pattern for three different tilt angles (0°, 45.4°, 90°).
Electronics 14 01404 g005
Figure 6. Graphical representation of placing two reader antennas.
Figure 6. Graphical representation of placing two reader antennas.
Electronics 14 01404 g006
Figure 7. Graphical representation of reader transmission flow (a) Conventional Pulse Protocol. (b) Modified Pulse Protocol.
Figure 7. Graphical representation of reader transmission flow (a) Conventional Pulse Protocol. (b) Modified Pulse Protocol.
Electronics 14 01404 g007aElectronics 14 01404 g007b
Figure 8. System throughput of conventional and proposed algorithm.
Figure 8. System throughput of conventional and proposed algorithm.
Electronics 14 01404 g008
Figure 9. System utilization of conventional and proposed algorithm.
Figure 9. System utilization of conventional and proposed algorithm.
Electronics 14 01404 g009
Figure 10. The execution time analysis of Modified Pulse protocol.
Figure 10. The execution time analysis of Modified Pulse protocol.
Electronics 14 01404 g010
Figure 11. Schematic representation of M-ary query trimming tree.
Figure 11. Schematic representation of M-ary query trimming tree.
Electronics 14 01404 g011
Figure 12. Time slot for identifying tag with (M = 2, 4, 8) for conventional and proposed algorithm.
Figure 12. Time slot for identifying tag with (M = 2, 4, 8) for conventional and proposed algorithm.
Electronics 14 01404 g012
Figure 13. System Throughput of conventional and proposed algorithm with (M = 2, 4, 8).
Figure 13. System Throughput of conventional and proposed algorithm with (M = 2, 4, 8).
Electronics 14 01404 g013
Figure 14. Total time slot with varying no. of tags of deterministic algorithms.
Figure 14. Total time slot with varying no. of tags of deterministic algorithms.
Electronics 14 01404 g014
Figure 15. Throughput with varying no. of tags of deterministic algorithms.
Figure 15. Throughput with varying no. of tags of deterministic algorithms.
Electronics 14 01404 g015
Table 1. Simulation Parameters.
Table 1. Simulation Parameters.
ParametersDenotationValue
CWContention Window32
T b i Beacon Interval5000 μ s
T m i n Minimum waiting time3 ×   T b i
T m a x Maximum waiting time2 ×   T m i n
T r e a d Reading time4 s
T c o l l i s i o n Collision beacon decrement interval2 ×   T b i
T s u c e s s Successful beacon decrement interval T r e a d + T m a x + T b i
T e m p t y Empty beacon decrement interval1 ×   T b i
l b Query transmission time 265   μ s
l q Beacon transmission time 341   μ s
Table 2. Comparison of algorithms to prevent tag collision.
Table 2. Comparison of algorithms to prevent tag collision.
TagsProtocolTotal Time SlotThroughput
100Binary tree132915%
Query tree664.450%
Collision tree564.466.6%
Proposed tree101.898.26%
200Binary tree305813%
Query tree152950.13%
Collision tree132966.6%
Proposed tree203.698.26%
300Binary tree493712.15%
Query tree246950.08%
Collision tree216966.6%
Proposed tree305.398.25%
400Binary tree691511.2%
Query tree345850%
Collision tree305866.6%
Proposed tree407.198.26%
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Mazumder, K.; Ghosh, A.; Singh, P.; Peddakrishna, S.; Kumar, J. RFID Antenna Deployment Model and Anti-Collision Algorithms to Enhance Vehicle Toll System Efficiency. Electronics 2025, 14, 1404. https://doi.org/10.3390/electronics14071404

AMA Style

Mazumder K, Ghosh A, Singh P, Peddakrishna S, Kumar J. RFID Antenna Deployment Model and Anti-Collision Algorithms to Enhance Vehicle Toll System Efficiency. Electronics. 2025; 14(7):1404. https://doi.org/10.3390/electronics14071404

Chicago/Turabian Style

Mazumder, Krishna, Anumoy Ghosh, Priyanka Singh, Samineni Peddakrishna, and Jayendra Kumar. 2025. "RFID Antenna Deployment Model and Anti-Collision Algorithms to Enhance Vehicle Toll System Efficiency" Electronics 14, no. 7: 1404. https://doi.org/10.3390/electronics14071404

APA Style

Mazumder, K., Ghosh, A., Singh, P., Peddakrishna, S., & Kumar, J. (2025). RFID Antenna Deployment Model and Anti-Collision Algorithms to Enhance Vehicle Toll System Efficiency. Electronics, 14(7), 1404. https://doi.org/10.3390/electronics14071404

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop