Multi-Carrier Spread-Spectrum & Related Topics: Third International Workshop, September 26-28, 2001, Oberpfafenhofen, Germany / Edition 1 available in Hardcover, Paperback
Multi-Carrier Spread-Spectrum & Related Topics: Third International Workshop, September 26-28, 2001, Oberpfafenhofen, Germany / Edition 1
- ISBN-10:
- 0792376536
- ISBN-13:
- 9780792376538
- Pub. Date:
- 02/28/2002
- Publisher:
- Springer US
- ISBN-10:
- 0792376536
- ISBN-13:
- 9780792376538
- Pub. Date:
- 02/28/2002
- Publisher:
- Springer US
Multi-Carrier Spread-Spectrum & Related Topics: Third International Workshop, September 26-28, 2001, Oberpfafenhofen, Germany / Edition 1
Buy New
$109.99-
SHIP THIS ITEMIn stock. Ships in 1-2 days.PICK UP IN STORE
Your local store may have stock of this item.
Available within 2 business hours
Overview
Product Details
| ISBN-13: | 9780792376538 |
|---|---|
| Publisher: | Springer US |
| Publication date: | 02/28/2002 |
| Edition description: | 2002 |
| Pages: | 376 |
| Product dimensions: | 6.10(w) x 9.25(h) x (d) |
About the Author
Read an Excerpt
On the improvement of HF Multicarrier modem behavior by using Spread Spectrum techniques to combat fading channels
Keywords: HF modem, Multicarrier Spreading, Multistage Interference cancellation
Abstract: This paper deals with the application of spread spectrum techniques to Multicarrier HF modems. It is observed that Multicarrier transmission over fading channels with deep nulls show a quite different performance between carriers: those in the vicinity of deep nulls provide a much higher Bit Error Rate (BER) than others far away of the nulls. Our proposal intends to spread different symbols over all the subcarriers simultaneously in order to homogenize performances. This approach, performed by orthogonal codes, shows an evident link with MC-CDMA techniques. In addition, joint detection strategies as Interference Cancellation (IC) with coherent detection are also applied in order to improve its performance. Several computer simulations support the feasibility of our proposal.
1. Introduction
HF transmission through ionospheric refraction is generally accepted as a very difficult task due to the time variant high dispersive channel characteristic. The ionosphere is a dispersive medium that spreads the signal in both time and frequency. In addition, received radio signals have usually been reflected from more than one ionosphere layer, a phenomenon known as multipath propagation. On the other hand, the ionization characteristic of this atmosphere layer also depends on solar conditions, season, latitude and local weather and time are in fact which the responsible of the time variant characteristic of this medium [1].
This behavior of the physical medium constrains the maximum feasible bandwidth to 3 KHz. Transmission rate over 1.8-2.4 Kbps. requires enhanced techniques with also complexity increase.
Most of the current modems are based on single carrier modulation including very powerful codes and long interleavers to compensate long burst errors [ 1 ]. Probably, the common basis of these modems is the Military Standard MIL-STD-188-110A [2] which specifies a continual transmission system to the maximum of 2400 bps.
One of the main civil applications of these modems is the HFDL system (High Frequency Data Link) developed by ICAO (International Civil Aviation Organization). HFDL is a TDMA system dealing with data transmission between aircraft and ground stations providing a data link by HF propagation meanwhile the airplane is far away the coast line [3]. Although this is a burst transmission instead of the continuous stream, the physical layer is almost identical to the single carrier modem of the MIL-STD-188-110A. In the current standard [3], long delays is not a critical point because it is devoted only for data transmission; even, if one slot is lost due to the improper behavior of the equalizer+interleaver+decoder, a request strategy provide the retransmission in probably better conditions. By our own, we are working in this application, but with a much wider purpose: we are focused on the capability of the system to carry out data and digital voice signals in an interactive operation mode. This approach requires a minimum delay also with data transmission rate over 2400 bps. These features do not apply for single carrier techniques but can be obtained by multicarrier modulation (OFDM) due to its more robust behavior against frequency selective fading channels [4, 5,6].
Let us describe briefly our proposal for the next generation of HFDL system in the next section. After that, we will focus on a particular improvement related with the use of multicarrier spreading; in fact, our approach can be formulated as a synchronized MC-CDMA where all the `users' must be demodulated.
2. The MC-HFDL System
This section is devoted to the presentation of the block diagram of the HFDL system. We will like to point out that the multicarrier version MC-HFDL is not a current standard but it has been presented at the ICAO panel on April 1999 at Montreal (Canada) as a potential alternative for the next definition of the standard [5]. Let us also remark the most critical constraints in our design:
- Our application is concerned with interactive communications where longer delays of 50-80 msec. are not acceptable.- We will incorporate a digital voice coder: of course, the maximum data rate achievable will provide the better voice quality. We expect to guarantee data rate up to 3600 bps even in bad channel conditions.
Figure 1 shows the block diagram of the OFDM data modem where all the main devices are presented both in transmission and reception.
We have labeled it a HF Spread-Spectrum OFDM modem. Due to space reasons, details about channel coding (Reed-Solomon (45, 63)), interleaving (frequency domain), time and frequency estimation schemes, pilot insertion, channel estimation ...must be referred to previous works [5, 6, 7]. On the other hand, spreader/despreader and interference cancellation is the main purpose of the present paper and will be discussed in detail in the next section.
Figure 2 shows an schematic view of the data structure to clarify the previous explanations. In this case, a pilot 16-OFDM symbol is inserted between couples of data 16-OFDM symbols just as an example. Data symbols of 32-OFDM and 64-OFDM are also under analysis. Also, the period insertion of the pilot symbol must be also determined; our experience shows that the channel variability requires channel estimation each 250-300 msec.
3. The Multicarrier Spreading
Let us now discuss about the application of the multicarrier spreading technique to the HF transmission. First, we will address the motivation and intuitive approach that may recommend the spreading. After that, a formulation of our approach as a problem of synchronized multiple access will link both disciplines: HF-OFDM and MC-CDMA. Finally, some computer simulations will confirm the significant improvement of the spreading technique.
Channel estimation can be easily accomplished by known pilots insertion. Several pilot patterns can be proposed: rectangular, hexagonal... which show a satisfactory performance in terms of its density and the final estimation error. In our application, delay is a critical point that unable pilot patterns employing several OFDM symbols. By our experience, the insertion of a pilot patter only in the frequency dimension allows a delay-free estimation with a computational very low cost by using the FFT algorithm. In fact, pilot symbol may use shorter time duration using alternate carriers and the final estimation for all the carriers is performed by interpolation.
However, in spite of the flat fading characteristic, a severe performance degradation can be observed at the subcarriers at the vicinity of deep nulls. Figure 3 shows a schematic view of the estimation procedure remarking the fact that at high SNR subcarrier, proper channel estimation is obtained and therefore an accurate demodulation is provided. On the other hand, at deep nulls, the local SNR decreases and coarse estimation may decrease the system performance dramatically. This is a typical observation in HF multicarrier modems: several carriers provide a low BER meanwhile other ones present a much higher BER which definitely decreases the expected performance...