Adaptive Forwarding Strategies in Content-Centric Networking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Named-data networks (NDNs) face challenges in content placement and dissemination due to dynamic and unpredictable content popularity, especially in ad-hoc wireless networks with mobility changes and varying air-link quality, which complicates the determination of optimal content retrieval routes.
Innovation Solution
The implementation of a strategic data forwarding layer using opportunistic intelligence and algorithms that probe the performance of multiple faces to select the best path for content retrieval, balancing delay and network conditions, while minimizing redundant data transmission and adapting to changing network conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If traditional IP host-to-host model is used for content delivery, then network infrastructure is simple, but it cannot support one-to-many and many-to-many delivery patterns required for content-centric networking
Solution Approach 1:
The patent introduces content routers as intermediary nodes that maintain forwarding information bases (FIB) and pending interest tables (PIT) to enable content-centric networking. These intermediaries handle content name-based routing, allowing one-to-many and many-to-many delivery patterns while abstracting the complexity from end devices. The content router acts as a mediator between content sources and consumers, implementing the named-data networking paradigm.
2Speed
If content is cached on network nodes to improve delivery speed, then user experience improves, but network overhead and memory requirements increase
Solution Approach 1:
The patent implements caching at specific network nodes (content routers) rather than universally across all nodes. Each router maintains a content store with cached content objects, allowing local delivery of popular content while avoiding unnecessary caching at nodes where it would consume resources without benefit. This localized caching approach optimizes the balance between delivery speed and resource consumption.
Solution Approach 2:
The system caches content partially - only when beneficial for future requests. The content store caches content objects based on demand patterns, caching frequently requested content while avoiding caching of rarely accessed content that would consume memory without providing performance benefits. This selective caching strategy optimizes the trade-off between delivery speed and network overhead.
3Adaptability or versatility
If multiple faces are probed to determine optimal routing path, then routing adaptability improves, but network overhead and delay increase
Solution Approach 1:
The patent implements probe interests that are sent in advance to explore multiple faces and determine their performance characteristics. By proactively probing faces before actual content requests, the system builds knowledge about face performance (latency, bandwidth, reliability) that can be used for subsequent routing decisions. This preliminary exploration reduces the need for reactive probing and minimizes routing delay for actual content delivery.
Solution Approach 2:
The system uses feedback from probe interests to update face performance metrics and selection probabilities. When probe interests are sent through different faces, the responses provide feedback about face quality, which is used to update the forwarding information base and adjust future routing decisions. This feedback mechanism enables adaptive routing that learns from experience and optimizes path selection over time.
4Measurement precision
If selection probability is updated based on probe feedback, then forwarding accuracy improves, but processing complexity and memory requirements increase
Solution Approach 1:
The patent changes the parameter representation from complex multi-dimensional performance metrics to simplified selection probabilities. Instead of tracking multiple performance parameters (latency, bandwidth, packet loss) separately, the system aggregates this information into a single selection probability value for each face. This parameter transformation simplifies the forwarding logic while maintaining measurement precision, as the probability values encapsulate the essential routing decisions.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A named-data networking (NDN) node, comprising a plurality of faces each of which is coupled to a different node in a content-centric network, and a processor coupled to the faces, wherein the processor is configured to probe the performance of each of the faces for an interest, wherein the interest is associated with multiple ones of the faces, wherein a next-hop is identified by the face, wherein a one of the faces associated with the interest is used to forward the interest when the interest is received by the NDN node, and wherein the one of the faces used is determined based on a selection probability determined from feedback from the probe of the performance of the faces.