Error detection/correction and fault detection/recovery – Pulse or data error handling – Data formatting to improve error detection correction...
Reexamination Certificate
2000-02-29
2001-06-12
Chung, Phung M. (Department: 2784)
Error detection/correction and fault detection/recovery
Pulse or data error handling
Data formatting to improve error detection correction...
C714S748000, C370S337000
Reexamination Certificate
active
06247150
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to transmission of data packets in a telecommunications network in which the transmitting end encodes the data blocks, divides them into smaller parts for interleaving, and transmits the parts in the form of bursts to the radio path, and in which the receiving end receives the bursts and deinterleaves them to reconstruct the original coded block. After the receiving end has received a faulty data packet, it requests the transmitting end to retransmit the packet. The retransmission protocol is ARC (Automatic Repeat Request).
BACKGROUND OF THE INVENTION
The current mobile networks enable bidirectional speech transmission between two parties, each of which can be a subscriber in a mobile network when the call travels within the same mobile network or through a circuit-switched PSTN/ISDN network from one mobile network to another. One of the subscribers can also be a subscriber in a circuit-switched PSTN/ISDN network. In any case, the connection is always circuit-switched and it is reserved for the use of these two parties for the entire duration of the data transmission. The known GSM system is a good example of a circuit-switched mobile network.
The mobile network was originally designed for effective speech transmission, and in current networks the data transmission rates are indeed quite low. Recently, mobile station users have been afforded an opportunity to attach to a packet-switched internet network through a circuit-switched mobile network. The terminal equipment may be the actual mobile station, comprising the suitable software, or the terminal may also be a computer attached to the mobile station, in which case the mobile station . . . use for speech transmission. Such cordless data transmission is attended by the drawback of slow data transmission, as GSM only offers the rate 9.6 kbit/s for data transmission.
This situation is improved by the GPRS (General Packet Radio Service) system using virtual circuits, which is currently being specified by ETSI (European Telecommunications Standards Institute). The purpose of GPRS services is to operate independently of the present circuit-switched services and particularly to utilize the unused resources of circuit-switched traffic. The GPRS system partly uses the Internet protocols, and hence a GPRS network can be directly connected to the Internet. The system has been logically implemented by superimposing it on the GSM system, adding two new network elements. The mobile station can be made bifunctional in such a way that it can serve as a normal GSM phone and as a GPRS phone relaying packet data.
Since both in a packet-switched mobile network and in a circuit-switched mobile network the links between mobile stations and base stations are radio links, the links are suspect to similar interference.
Simultaneous connections cause mutual interference the magnitude of which is dependent on the channels used for the connection, the geographical location of the connections, and the transmission power employed. These can be influenced by planned channel allocation to different cells which takes interference into account, and by transmission power control. The distance at which the same channel can be reused while the signal carrier to interference ratio (CIR, C/I) remains acceptable is called the noise distance.
FIG. 1
illustrates the effect of an interference signal at the reception. A burst signal (wanted signal) is anticipated to arrive in a reception time slot. At some phase thereof, often in the middle, the signal comprises a training sequence known to the receiver, in accordance with which the receiver adjusts its channel corrector. If an interference signal of the same frequency arrives simultaneously, it destroys the wanted signal entirely or at least partly. If the interfering signal arrives with a delay as in the figure, it is nearly impossible to detect the bits at the end of the wanted signal. If part of the interfering signal arrives simultaneously as the part comprising the training sequence is being received, the receive signal is completely lost. The interfering signal can be a multipath-propagated component of the same transmission, or it can be a signal originating from a different source but arriving at the frequency of the wanted signal.
Since the use of a higher CIR ratio than necessary in digital systems hardly improves connection quality, the transmission power used on the connections is dynamically controlled. The requisite power is dependent on channel fading between the mobile station and the base station, the interference caused by other connections, and ambient noise. Interference can also be diminished for example by using directional antennas, in which case the same signal level can be achieved at the receiver with lower transmission power.
Also Doppler shift causes interference in transmission. The frequency change produced thereby causes rotation of the received burst and impairs the accuracy of the channel estimate, calculated on the basis of the training sequence located in the middle of the burst, towards the end of the burst. This is illustrated in
FIG. 2
, in which the signal
oise ratio is good in the middle of the burst but deteriorates at the beginning and at the end.
In addition to the CIR representing the radio channel quality, the connection quality is influenced by the sensitivity of the information signal transferred over the channel to transmission errors arising in the radio channel. The information can be rendered more immune to transmission errors by processing the information prior to its transmission to the channel by channel coding and interleaving and by using re-transmission of faulty data frames.
This is illustrated in FIG.
3
. In accordance with the figure, the transmitting end channel codes the transmit data in blocks, splits the blocks into smaller parts and changes the order of the parts (interleaving). Thereafter the data is transmitted in bursts through the radio interface to the receiving end, which performs the same operations in reverse order.
The purpose of channel coding is on the one hand to render the information transfer more immune to transmission interference and on the other hand to detect transmission errors. In channel coding, redundancy by means of which errors caused by the radio channel can be corrected and non-correctable errors detected at the signal receiving end is added to the actual user data to be transmitted. Whilst affording better interference immunity, channel coding increases the bandwidth requirement for information transfer.
The bit errors produced in the radio path are typically error bursts comprising a sequence of several bits. Individual bit errors are always easier to correct than a sequence of several successive erroneous bits. The probability of several successive erroneous bits occurring can be significantly reduced by bit interleaving, in which the order of the bits is scrambled in a predetermined manner prior to the sending of the signal to the radio path. When the relative order of the bits is restored to original at the receiving end, the bits in which radio path interference has caused errors are no longer adjacent, and thus the errors are easier to detect and correct. Whilst affording enhanced error correction and detection, interleaving produces a slight additional delay in the data transmission.
By using stronger channel coding and deeper interleaving, the user data can be transported to the receiver in a sufficiently error-free state even over a radio channel that is poorer than normal. Power control, interleaving and coding are the conventionally used means for correcting burst errors resulting from fading, interference and Doppler shift. In speech transmission, these measures are sufficient, as any small number of lost speech frames are replaced at the receiving end by constructing replacement frames in which the previously received speech parameters are utilized. In packet-switched networks which transfer mainly data records, these methods do not as such afford a suff
Altera Law Group LLC
Chung Phung M.
Nokia Networks Oy
LandOfFree
Automatic retransmission with order of information changed does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Automatic retransmission with order of information changed, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Automatic retransmission with order of information changed will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2517030