Barrage Relay Network – Barrage Relay, Cooperation and BAC

System Initializing…

TDMA Frame 1

TDMA Frame 2

Barrage Relay

How data floods a Barrage Relay Network slot-by-slot

TDMA Frame 1 • Slot A

The Initial Broadcast

The central Source node (black) transmits the first packet. This is visualized as the first expanding blue wave. All nodes that successfully receive this packet are defined as being one hop away from the source.

TDMA Frame 1 • Slot B

First Cooperative Relay

All 1-hop nodes simultaneously transmit the exact same packet they just received. Notice how multiple nodes send overlapping waves to the 2-hop receivers. Because of autonomous cooperative communication (phase dithering), these packets neither collide nor result in destructive interference.

TDMA Frame 1 • Slot C

Second Cooperative Relay

The ripple continues outward. Now, the nodes that are two hops away simultaneously transmit the identical information to nodes further out on the network edge. The original packet is smoothly flowing away from the source.

TDMA Frame 2 • Slot A

Spatial Pipelining (The Magic Trick)

Two things happen simultaneously here! The 3-hop nodes relay the original blue packet. Because those 3-hop nodes are far enough away, the 1-hop nodes will not hear them. This spatial reuse allows the Source to safely transmit a brand new packet (the green wave) into the network without causing a collision.

Autonomous Cooperation

Phase Dithering & Blind Composite Processing

Tx1Teammate 1

Tx2Teammate 2

Tx3Teammate 3

RxBlind Receiver

✓

1. Simultaneous Transmit Nodes Tx1, Tx2, and Tx3 send identical packets on the exact same TDMA slot (PHY-Layer Switching).

2. Phase Dithering The dashed lines show how each node randomly shifts its phase, creating time-diversity instead of destroying the signal.

3. Blind Processing The Receiver uses Error Correction to piece the overlapping signals together into a perfect packet (✓) without knowing who sent it.

Barrage Access Control (BAC)

Traditional networks use MAC (Media Access Control) protocols to manage traffic across single, point-to-point links. Because Barrage Relay Networks (BRNs) flood the entire network at once, standard link-based protocols don’t work. If multiple nodes try to broadcast massive amounts of data simultaneously, the network will crash.

To solve this, BRNs use Barrage Access Control (BAC). BAC acts as a network-wide traffic cop, granting temporary ownership of the entire broadcast fabric to a single source at a time. It achieves this by dividing the network’s time slots into three distinct logical channels.

Key Steps of the BAC Protocol

Request (RLC – Request Logical Channel) When a node has data to send, it broadcasts a request message on the RLC. If multiple nodes request at the exact same time, a collision happens here.

Confirmation (CLC – Confirmation Logical Channel) A dedicated Schedule Control Node listens to the requests. It creates a schedule of who gets to transmit and when, and broadcasts this schedule over the CLC. Because there is only one scheduler, collisions never happen on the CLC. (If a node sent a request but doesn’t hear its name on the CLC, it assumes its request collided and tries again later).

Data Transmission (DLC – Data Logical Channel) Once a node is confirmed on the schedule, it becomes the “designated source.” It injects its data packets into the DLC, and the rest of the network relays that data using the standard barrage broadcast method.

Barrage Access Control (BAC)

How Barrage Relay Networks safely manage multiple data sources without crashing the network.

RLC

Request Channel

A node broadcasts a request indicating it wants to send data. If multiple nodes ask at once, collisions happen here.

CLC

Confirmation Channel

The centralized Schedule Control Node updates the network schedule and broadcasts who gets to transmit next.

DLC

Data Channel

The approved node takes control of the network fabric and floods its data payload to all connected nodes.

Reference

[1] T. R. Halford and K. M. Chugg, “Barrage Relay Networks,” in 2010 Information Theory and Applications Workshop (ITA), Jan. 2010, pp. 1–8. doi: 10.1109/ITA.2010.5454129.