 object mapping is the process of translating the identity at the higher layer to the more tangible commonly known addressing at the lower layers. So we need to understand what are the identifiers and what are the options for identifying objects at various layers then what is what are the mapping requirements and what could be the possible options the identifier is basically the identity of an entity at a certain layer so a physical host like a computer could have multiple or many many objects or IDs associated with it each one requiring some unique identifier IPv6 ngn addressing for different types of objects requires each object to have connectivity to the network as well as globally recognizable and contactable attributes so unique object ID is crucial for end-to-end communication and we know that in internet of things and web of things each entity as small as a Fitbit or perhaps a user app needs to have an identity so we start off with very basic concepts of identifiers like URLs, URIs, the E.164 numbering scheme and the classical IP and MAC addressing let's now look at the mapping requirements the mapping requirements directly come out of the relationship between the objects their identifiers and the identities which are translated into more well-known addresses at the transport stratum so we have the application we have the applications where each user an object has to have some identities now these identities need to be given IDs and these IDs have to be translated into addresses the relationship could also be seen as names getting translated into URLs getting translated into MAC IPv6 addresses session IDs etc so the requirements which are going to come out of it are first of all the network layer has to provide separation of the node identity and the locator roles because we know that if you look at a IPv4 address in particular it has two purposes the first one is it identifies the location where this particular IPv4 address is situated then it also identifies the object or the host to which it is binded so we have certain protocols like the host identity protocol which performs this decoupling between the ID and the locator purpose of the IP addresses by using the identity tags it has its own implications it makes things more complex then we have a certain mapping options like we could have direct mapping or we could have indirect mapping direct mapping is where the object is directly located in the end host via the same point of attachment but in other cases an object could only be accessible through a gateway or through an ad care of address as in mobile IP version 6 so the mapping options which could possibly be realized could be that mapping could be one-to-one when the host itself is the object so we could have a computer or a telephone a simple example like that but we know that one host could have many objects so we need to have mapping between host and objects so in this case host is not the object but host has many objects so we need to provide the virtual and physical content based objects with own identifiers for that we need the object mapping that we've just discussed sometimes a gateway node maintains an object gateway table for mappings now this is a unique thing which could also be realized where we dedicate an entity known as the gateway that keeps track of all the objects which reside or are contained within a host