DE102020101231A1 - Method for encoding and decoding packets in random access protocols - Google Patents

Abstract

The present invention proposes a coding scheme in which the transmission device also supplies part of information by selecting the slot in which it transmits. If a packet is detected in a special slot on the recipient side, the recipient receives this information at no additional cost.

Description

Technical area

The invention relates to a method for coding and decoding packets in random access protocols in the field of data transmission and, in particular, to a data transmission method using coded-slotted Aloha schemes, a transmission device and a receiving device.

background

It is very important that it is possible to reliably transmit and receive data. In a slotted random access scheme in which slots are grouped in Medium Access Control (MAC) frames, they all have the same length N (in slots). Each slot consists of n channel usages, which leads to a total of nN channel usages for a single MAC frame. Each user is frame and slot synchronous and tries to transmit a maximum of one packet per MAC frame. Disregarding guard intervals, the duration of a packet is n. At the beginning of a MAC frame, each user generates a packet to be transmitted in the frame with a probability π, where π is called the activation probability. Users who attempt to transmit within a MAC frame are referred to as active users for that frame.

The conventional method, for example the method described by E. Paolini, G. Liva and M. Chiani “Coded slotted aloha: a graph-based method for uncoordinated multiple access” works as follows: it is assumed that an active user has received a message from k bits to transmit. The k-bit information vector u is coded by the user via a binary linear block code with a generator matrix G, which generates a code word c with a length n, i.e. c = uG, which leads to a code rate of R = k / n. For each transmission the user generates an arbitrary index i ∈ {1,2, ...., N} and transmits his packet c in slot i. With each transfer, each user creates his index uniformly and independently of all previously selected indices and without coordination with the other users. If there is no collision, i.e. there is no more than one user transmitting his packet in the same slot, the receiver tries to estimate the message u using the noisy observations y. Of course, an attempt can be made to optimize the distribution π in order to eliminate the collisions or to handle them in such a way that successive interference cancellation (SIC) helps to significantly reduce the error probability. We are interested in coding schemes with a low complexity physical layer in order to improve the reliability, i.e. to reduce the probability of an erroneous estimate ü whenever there is no collision in a slot.

The problem addressed in the present invention is therefore how the decoding probability can be improved in coded-slotted-aloha schemes in which several users transmit their messages to one recipient in an uncoordinated manner.

This task is accomplished with the features of claims 1 and 5 solved.

The present disclosure provides a method of transmitting data, a transmitting device, and a receiving device, and the technical solutions provided by the embodiments of the present disclosure can improve the decoding probability.

A method for transmitting data is provided in which a transmission channel to a receiving device is shared by a plurality of users, the data being transmitted using the Coded Slotted Aloha, CSA, scheme, in which an information vector of the transmitted Data is divided into two parts, the slot indices in a medium access control, MAC, frame being selected exclusively by a first part and only a second part being encoded by a transmission device, the receiving device being the first part of the information vector by detecting the slot and decoding the second part of the information vector.

In the conventional methods, the information that can be transmitted thanks to the slot index selected by a user (i.e. discrete timing information) is not used to transmit the message. In other words, it is possible to embed a fraction of the information free of charge in the slot in which the information is transmitted to improve performance. In other words, the first part of the vector still contains data which, however, is not encoded but is transmitted with the selected index. Therefore, the index is not selected according to a different algorithm, but the first part of the data vector is used to select the index, so that this part of the information vector can be transmitted free of charge.

Preferably, the method is further with a successive interference suppression, SIC, method or with coding schemes with a one low complexity physical layer combined.

Each MAC frame preferably has N slots, the information vector u = (u _i , u _d ) k bits, where u _{i is} the first part with a length m = log ₂ ^N , the second part u _d of a (n, k ') channel code is encoded with k' = km.

The estimate of u _i is preferably based on the one-to-one mapping between u _i and the slot index, u _{d is} decoded with the decoder of the (n, k ') channel code.

According to a second aspect of the embodiments of the present disclosure, a transmission device is provided which is designed such that it divides an information vector into two parts, the slot indices in a medium access control, MAC, frame exclusively from a first part is selected and only a second part is encoded by the transmission device.

Each MAC frame preferably has N slots, the information vector u = (u _i , u _d ) k bits, where u _{i is} the first part with a length m = log ₂ ^N , the second part u _d of a (n, k ') channel code is encoded with k' = km.

According to a third aspect of the embodiments of the present disclosure, a receiving device is provided which is designed such that it receives a two-part information vector, a first part of the information vector being estimated by detecting the slot and a second part of the information vector being decoded by a decoder.

A preferred embodiment of the invention is described below with reference to the figure.

The estimation of the first part is preferably based on the one-to-one mapping between the first part and the slot index; the second part is decoded with the corresponding decoder.

The present invention proposes a coding scheme in which the transmission device also supplies part of information by selecting the slot in which it transmits. If a packet is detected in a special slot on the recipient side, the recipient receives this information at no additional cost.

1. 1. Teilen eines k-Bit-Informationsvektors u in zwei Teile als u=(u_i, u_d), wobei u_i ein Vektor mit einer Länge m ist, wobei m=log₂ ^N ist, das den von der Übertragungseinrichtung gewählten Slot-Index angibt.
2. 2. Die Übertragungseinrichtung verwendet einen (n,k') Kanalkode mit k'=k-m zum Kodieren des Vektors u_d mit einer Länge k'.
3. 3. Nach dem Detektieren des Slot eines Pakets am Empfänger erfolgt die Schätzung von u_i kostenlos, da es ein Eins-zu-Eins-Mapping zwischen u_i und dem Slot-Index gibt. Mit anderen Worten wird der Teil u_i durch Detektieren des Slot geschätzt.
4. 4. Dann dekodiert er den empfangenen Vektor mit dem Dekodierer des (n,k') Kodes zum Dekodieren von u_d.

In particular, a Coded Slotted ALOHA (CSA) scheme with N slots comprises the following steps:

1. 1. Divide a k-bit information vector u into two parts as u = (u _i , u _d ), where u _{i is} a vector of length m, where m = log ₂ ^N , which is the slot selected by the transmission device -Index indicates.
2. 2. The transmission device uses an (n, k ') channel code with k' = km for coding the vector u _d with a length k '.
3. 3. After the slot of a packet has been detected at the receiver, u _i is estimated free of charge, since there is a one-to-one mapping between u _i and the slot index. In other words, the part u _i is estimated by detecting the slot.
4. 4. Then it decodes the received vector with the decoder of the (n, k ') code to decode u _d .

At the transmission facility, the message is divided into two parts as u = (u _i , u _d ), where u _{i is} a vector of length m that selects the index in a MAC frame, and u _{d is} the vector of length k 'which contains the other part of the message. Then, a transmission device encodes only the part u _d using an (n, k ') code to transmit the k-bit information vector u.

At the receiver the estimated part û _{i is} only recovered by detecting the packet if there is no collision in the slot. What remains is to retrieve the estimated û _d for the second part using the decoder of the channel code.

In particular, such a scheme can be combined with the SIC method.

1 shows the performance of the method of the invention.

The performance increase for such a coded-slotted ALOHA scheme with N = {128, 256, 512,1024} due to the assumption that each slot has n = 128 channel users is based on the assumption that 64-bit information for one Channel code are transmitted, which leads to a code rate of R = 0.5. In the proposed scheme, m = {7, 8, 9, 10} bits can be transmitted using the discrete timing information thanks to the slots, which means that the channel code used is k '= {57, 56, 55, 54, respectively } Has.

A very tight finite length coding bound called a Random Coding Union (RCU) bound is provided, which is roughly the performance of the best coding scheme for given parameters. The increases in the proposed scheme are in 1 specified.

The proposed method can be used with all types of commercially available wireless transmission systems.

Claims (8)

Method for transmitting data, wherein a transmission channel to a receiving device is shared by a plurality of users, wherein the data is transmitted using a Coded-Slotted-Aloha, CSA, scheme, An information vector of the data to be transmitted is divided into two parts, the slot indices in a medium access control, MAC, frame being selected exclusively by a first part and only a second part being coded by a transmission device, wherein the receiving device estimates the first part of the information vector by detecting the slot and decodes the second part of the information vector. Procedure according to Claim 1 , the method being combined with the successive interference suppression, SIC, method or coding schemes with a low complexity physical layer. Method according to one of the Claims 1 or 2 , in which each MAC frame has N slots, the information vector u = (u _i , u _d ) k bits, where ui) is the first part with a length m = log ₂ ^N , the second part u _d of a ( n, k ') channel code is encoded with k' = km. Procedure according to Claim 3 , in which the estimate of u _{i is} based on the one-to-one mapping between u _i and the slot index, u _{d is} decoded with the decoder of the (n, k ') channel code. Transmission device which is designed such that it divides an information vector into two parts, the slot indices in a medium access control, MAC, frame being selected exclusively by a first part and only a second part being encoded by the transmission device . Transfer device according to Claim 5 , in which each MAC frame has N slots, the information vector u = (u _i , u _d ) k bits, where u _{i is} the first part with a length m = log ₂ ^N , the second part u _d of a ( n, k ') channel code is encoded with k' = km. Receiving device which is designed such that it receives a two-part information vector, a first part of the information vector being estimated by detecting the slot and a second part of the information vector being decoded by a decoder. Receiving device according to Claim 7 , in which the estimation of the first part is based on the one-to-one mapping between the first part and the slot index, the second part is decoded with the corresponding decoder.

Images